基于双激光标靶图像识别的掘进机位姿检测方法

发布时间：2018-03-07 12:31

本文选题：测量　切入点：位姿检测　出处：《激光与光电子学进展》2017年04期 　论文类型：期刊论文

【摘要】：针对煤矿井下巷道掘进机位姿的测量需求,采用双激光标靶的图像识别测量方式,构建了位姿实时检测系统。研究了激光标靶上光学特征点和特征光线的空间分布,提出一种基于双激光标靶的掘进机位姿解算方法。通过对图像数字处理得到精确的特征点、特征光线参数,实现对标靶参考点的精确定位,使用空间矩阵变换方法,解算出掘进机位姿参数。在实验室条件下建立机身位姿自动检测实验平台,模拟环境下的测试结果表明,在测量范围小于40m时,位移误差小于2mm,角度误差小于0.5°。该系统测量精度高,结构简单可靠,实时性强,能够很好地满足测量需求,实现对掘进机在煤矿井下掘进过程中的位姿检测。
[Abstract]:According to the requirement of position and pose measurement of roadway tunneling machine in underground coal mine, a real time detecting system of position and pose is constructed by using the image recognition and measurement method of double laser target. The spatial distribution of optical characteristic points and characteristic light on laser target is studied. In this paper, a method for calculating the position and pose of tunneling machine based on double laser target is proposed. The accurate feature points and characteristic ray parameters are obtained by digital image processing, and the precise location of target reference points is realized, and the spatial matrix transformation method is used. The position and attitude parameters of the tunneling machine are solved. An automatic testing platform for the fuselage position and pose is established under the laboratory conditions. The test results in the simulated environment show that, when the measuring range is less than 40 m, the test results show that, when the measurement range is less than 40 m, The displacement error is less than 2 mm and the angle error is less than 0.5 掳. The system has the advantages of high measuring precision, simple and reliable structure and strong real time performance.
【作者单位】：中国矿业大学信息与电气工程学院;
【分类号】：TD421.5;TP391.41

【相似文献】